1. Field of the Invention
This invention relates to methods of identifying printers, more particularly to methods for controlling the print output to allow identification.
2. Background of the Invention
Watermarking on printed paper gives anyone receiving the paper a means to ensure that the paper and what it is printed on it is authentic. Examples include currency, stock certificates, money orders and government documents. With the advent of digital transmission of images to be printed, however, watermarking does not work as effectively as when most transactions involved paper.
Examples of this type of electronic watermarking are shown in U.S. Pat. No. 5,606,609, issued to Houser, et. al., on Feb. 25, 1997. In this patent, the data is marked with a security object that contains security information, some type of electronic signature and an identifier. The identifier invokes the security information when the document is displayed. The integrity of the document is then verified and the signature is displayed. If the integrity does not verify then the signature is not displayed.
Another example of this type of watermarking is shown in U.S. Pat. No. 5,488,664, issued to Shamir on Jan. 30, 1996. In this patent, a watermark is displayed with the image, thereby allowing visual authentication of the image. The watermark is incorporated by the method in one of two ways. One method prints groups of subpixels on two different layers, an original and a decoding sheet. The other method prints the subpixels on two layers of the original. Both layers must be displayed and viewed simultaneously to show the watermark.
However, all of these patents focus primarily on protecting the display of the image, rather than the printout of the image. Once the image is printed, in most cases, it is nearly impossible to tell from where it came.
One patent concentrates in this area, U.S. Pat. No. 5,568,550, issued to Ur, on Oct. 22, 1996. The patent is directed to identifying the software copy responsible for generating printed images.
In this patent, the method includes a dispersed image of dots in predetermined locations around a printed image. The dispersed image of dots is extracted from the printed image by a high-resolution scanner. The dispersed dots are put into the print image data at the print file, making them difficult to identify and extract the associated commands. However, even though this method tries to alleviate any problems in the resulting image quality, it is possible that it can adversely affect the quality. Further, the extraction of the pattern relies upon the clarity of the document and the availability of large areas of the printed page. This patent does not identify the print engine itself, just the software using the printer or print engine.
Identification of the printer or print engine has uses in security and law enforcement areas. If officials in these areas can determine from what printer a particular output was made, they can more effectively prevent unauthorized hard copies of secure documents and forgery of unauthorized documents, as examples.
Therefore, a method and apparatus is needed which allows identification of the printer by the output it produces, without having a detrimental effect on the print image quality.
An aspect of the invention is a method for inserting an identification modulation code into a printed image. The method comprises the steps of inserting a start code followed by the identification modulation code and then a stop code. The code is determined by the magnitude of the subpixel modulation. The magnitude of the subpixel modulation is decoded to determine the value of the code in that document.